Discussion of The Bank of England s Approach to Stress Testing the UK Banking System

Transcription

1 Philipp Hartmann European Central Bank Discussion of The Bank of England s Approach to Stress Testing the UK Banking System London 30 October 2015 London School of Economics/CEPR Conference on Stress Testing and Macro-prudential Regulation: A Trans-Atlantic Assessment Disclaimer: Any views expressed are only the speaker s own and should not be regarded as views of the ECB or the Eurosystem

2 Two versions of the macro-prudential policy objective Asymmetric Ensure systemic financial stability Avoid financial crises Prudential taken literally (ex ante perspective) Symmetric Ensure the smooth flow of financial intermediation services (ex ante and ex post perspective) Governing Council Smoothen the financial cycle, both in the upturn and in the downturn More like a third aggregate stabilisation policy (in addition to monetary and fiscal policy) More ambitious 1

3 Annual stress test with cyclical scenario Countercyclical design of stress test Upturn of financial cycle: harsher scenario Rapid growth of credit and asset prices Compressed risk premia Effective instability low but underlying risks growing (emerging imbalances?) Downturn of financial cycle: softer scenario Governing Council Correction of credit provision and asset prices High risk premia Effective instability large Some steps towards symmetric approach Rules versus discretion: Constrained discretion soft harsh harsh soft 2 t

4 Policy responses and challenges If the stress test results suggest changes of capital 1) Financial Policy Committee considers system-wide buffers 2) Prudential Regulation Authority considers individual banks buffers Question: Can PRA also reduce some individual buffers or creates the FPC a hard floor? Challenges How to determine the state of the financial cycle and be consistent over time? Indicators may point in different directions Judgement Design may work better in regular financial cycles than in crisis cycles: In a severe financial crisis it may sometimes be necessary to re-establish supervisory credibility and market confidence by generating a larger capital increase (e.g. if minimum requirements are violated) Probably many more Governing Council 3

5 Risk in a severe crisis: Illustration from 3D model Effects of a persistent reduction in home prices and firm valuations on GDP CCB CCB per cent per cent NO CCB NO CCB Governing -1.8 Council quarters Higher capital requirements (10.5%) attenuates shock propagation for the first 1.5 years quarters Lower capital requirements (8%) attenuates shock propagation early on But when capital becomes too low excessive defaults may also worsen the situation Source: Clerc, Derviz, Mendicino, Nikolov, Moyen, Stracca, Suarez and Vardoulakis (2015), Capital regulation in a macroeconomic model with three layers of default, International Journal of Central Banking, 11(3), pp Novel approach for assessing the benefits and costs of macro-prudential regulatory policies developed under the ESCB Macro-prudential Research Network (MaRs). 4

6 Other comments Capital policies may not be the most effective regulatory instrument against the sources of financial imbalances Evidence that borrower based instruments more effective than lender based instruments (Claessens, Gosh and Mihet 2014) Capital may not lean as much against the financial cycle as LTVs, DTIs and the like Did the FPC s secondary objective play a role in excluding (for the Governing moment) Council medium-sized banks? (box 1) Less burden on them Easier to innovate and grow to compete with the large ones in the future? Work with home supervisor that the parent group supports the UK investment banking subsidiary of a foreign-owned bank? (p. 24) We agree that it should be explored how stress tests can be extended to the wider financial system (box 5) 5